Diurnal vegetation moisture cycle in the Amazon and response to water stress

Diurnal Vegetation Moisture Cycle in the Amazon and Response to Water Stress Milad Asgarimehr 1,2,3 , Dara Entekhabi 4 , and Adriano Camps 1,5,6 1 Signal Theory and Communications Department, CommSensLab—UPC, Universitat Politecnica de Catalunya, Barcelona, Spain, 2 Institute of Geodesy and Geoinformation Science, Technische Universität Berlin, Berlin, Germany, 3 Section 1.1 Space Geodetic Techniques, German Research Centre for Geosciences GFZ, Potsdam, Germany, 4 Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA, 5 IEEC—Institut d’Estudis Espacials de Catalunya, Barcelona, Spain, 6 College of Engineering, UAE‐University, Al Ain, United Arab Emirates Abstract Water stress in the Amazon is exacerbated by rising temperatures and reduced moisture levels. However, understanding forest responses to increased aridity is hindered by limited in situ water potential observations in the Amazon. Remote sensing of water content has emerged as a promising metric. Vegetation Water Content (VWC) diurnal dynamics is hypothesized to reflect water stress responses. Conventional sensors' low sampling rates impede capturing and studying sub‐daily VWC dynamics. Leveraging Global Navigation Satellite System Reflectometry (GNSS‐R) with unprecedented sampling rates, this study reveals significant disparities in morning and evening VWCs in the Amazon, for example, by ≈1.1 and 1.0 kg/m2during the wet and dry seasons of 2019. A strong correlation (R=0.8)between ΔVWC (the difference between evening and morning VWCs) and vapor pressure deficit is observed in Amazonian peatland. This highlights the potential of VWC from innovative remote sensing techniques in elucidating water stress dynamics in critical ecosystems. Plain Language Summary In the Amazon rainforest, rising temperatures and decreasing moisture levels are causing plants to experience more water stress. However, scientists have struggled to understand how the forests are responding to these drier conditions as direct measurements of plant moisture content do not provide sufficient coverage. Recently, researchers have started using satellites to measure water in plants, which could help us understand how they are coping with the lack of water. However, conventional sensors hardly offer measurements often enough to capture the daily changes in plant water levels. This study uses a novel satellite observation technique, Global Navigation Satellite System Reflectometry, that offers measurements with unprecedented frequency. It is found that there are significant differences in plants' water content in the morning compared to the evening in the Amazon, for example, by ≈1.1 and 1.0 kg/m2during the wet and dry seasons of 2019. This study reveals that the difference level responds significantly to environmental aridity. As a result, novel satellite methods could help us better understand how water stress is affecting the Amazon rainforest. 1. Introduction The Amazonian rainforests wield significant influence within the climate system contributing to the exchange of energy, water, and carbon dioxide (CO2), while also serving as a pivotal carbon reservoir. However, it is anticipated that climate change will result in a considerable increase in temperature and a potential reduction in moisture levels within the Amazon forest throughout the 21st century (W. Yuan, Zheng, et al., 2019). Droughts diminish tree growth in old‐growth forests in the Amazon region (Hubau et al., 2020). The loss of biomass indicates that species within this area are functioning beyond their hydraulic thresholds (Tavares et al., 2023), and the occurrence of frequent drought events has led to an increase in tree mortality (Hartmann et al., 2022). Water availability profoundly influences photosynthesis and evapotranspiration processes (Y. Liu, Li, et al., 2020), posing a significant threat to plant survival and contributing to forest mortality. However, monitoring photosynthesis to assess water stress trends remains contentious, as the seasonal patterns of photosynthesis in Amazonian forests are closely tied to phenology (J. Wu et al., 2016) rather than directly indicating aridity. Despite diurnal patterns revealing a noteworthy depression in photosynthesis and evapotranspiration during dry season afternoons, a positive response to Vapor Pressure Deficit (VPD) is observed in the mornings (Z. Zhang et al., 2023). Therefore the relationship between these factors is not straightforward and generalizable due to the RESEARCH LETTER 10.1029/2024GL111462 Special Collection: Hydrogeodesy: Understanding changes in water resources using space geodetic observations Key Points: •Global Navigation Satellite System Refractometry offers unprecedented sampling, unveiling Amazon's diurnal vegetation water content •Vegetation water content generally peaks in mornings, fluctuating significantly throughout the day •Amazonian peatland's VWC diurnal cycle correlates strongly R=0.8 with vapor pressure deficit, proposed as a water stress indicator Supporting Information: Supporting Information may be found in the online version of this article. Correspondence to: M. Asgarimehr, [email protected] Citation: Asgarimehr, M., Entekhabi, D., & Camps, A. (2024). Diurnal vegetation moisture cycle in the Amazon and response to water stress. Geophysical Research Letters,51, e2024GL111462. https://doi.org/10.1029/ 2024GL111462 Received 5 AUG 2024 Accepted 15 SEP 2024 © 2024. The Author(s). This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. ASGARIMEHR ET AL. 1 of 11 heterogeneity in the Amazonian landscape vegetation types. This issue is further compounded by a lack of large‐ scale observations of photosynthesis or evapotranspiration within Amazon forests. Gradients in soil and plant water potential (ψ)obtained from in situ measurements have been considered as a direct indicators for gaining insights into vegetation water stress (T. J. Brodribb et al., 2020). However, progress in understanding the vegetation hydraulic mechanism under water stress in the Amazonian has been limited due to the discrete and sparse nature of in situ water potential observations (Novick et al., 2022). In situ and eddy covariance flux observations are rare especially in the Amazonian rainforests due to logistical challenges, harsh environmental conditions, limited infrastructure, hampering extensive and long‐term monitoring efforts in the region. Therefore dynamic global vegetation models struggle to accurately capture seasonal carbon exchanges in Amazonian tropical forests, particularly in representing dry‐season declines and the interplay between internal biophysical processes and environmental factors (Restrepo‐Coupe et al., 2017). Detecting and mapping water stress in Amazon rainforests using alternative approaches intensifies due to impending climate impacts. Vegetation Water Content (VWC), a novel remote sensing product, offers an alternative metric for assessing water status. This parameter refers to the amount of water contained within the vegetation, including water in the plant tissue and surface. When derived from microwave remote sensing, VWC is measured using the differences in the dielectric properties between water and dry vegetation and is normally expressed as the mass of water per unit ground area (e.g., kg water/m2). Recent studies on VWC using microwave sensors have demonstrated significant advancements and challenges (Konings et al., 2021). Rao et al. (2019) utilized Vegetation Optical Depth (VOD) from passive sensors, which is strongly (and often assumed linearly) linked to VWC through the microwave signal attenuation properties of water within vegetation, to assess drought‐driven tree mortality. Konkathi and Karthikeyan (2024) focused on the utility of L‐band and X‐band VOD derived from radiometer sensors for examining VWC and found that passive microwave remote sensing effectively represents VWC. Recently, Bernardino et al. (2024) employed SAR data to estimate VWC in Brazil using active sensors, revealing that savannas retain water better during dry seasons, thereby aiding in drought response and land management. However, challenges remain in effectively capturing short‐scale VWC dynamics to detect water stress in early stages. VWC exhibits diurnal fluctuations that are influenced by the processes of water uptake and transpiration in plants. During the daytime, plants lose water through transpiration. This loss of water leads to a decrease in VWC as the day progresses. At night, when photosynthesis and transpiration rates drop significantly, plants replenish their water content through root uptake from the soil. This nighttime uptake restores the VWC, leading to higher levels by early morning. Therefore, the diurnal VWC cycle becomes a key indicator of water stress when plants fail to replenish the water lost during the day (Nelson et al., 2018). Examining VWC at a diurnal scale enhances the detection of plant water stress and fosters a deeper understanding of forest‐weather interactions. These insights into vegetation diurnal processes elucidate the mechanisms governing the exchange of CO2and water vapor between the Amazon rainforest and the atmosphere, thereby revealing crucial aspects of ecosystem functioning (Xiao et al., 2021; W. Wang et al., 2011). Conventional radar sensors have shown sensitivity to sub‐daily VWC fluctuations (Konings et al., 2017; Paget et al., 2016; van Emmerik et al., 2017), but their sampling rates are hardly adequate for monitoring diurnal dynamics. Konings et al. (2021) state “our ability to monitor VWC dynamics with microwave remote sensing is currently constrained by sensor availability, not by technology.” However, with the emergence of novel remote sensing frameworks, addressing the issue of sensor availability becomes possible. 2. Methodology: GNSS Reflected Signals Carry VWC Information Recent studies suggest using Global Navigation Satellite System (GNSS) to assess vegetation moisture by analyzing direct signals received at ground stations (Camps, Alonso‐Arroyo, et al., 2020; Yao et al., 2024). Nonetheless, this approach falls short in providing adequate coverage for large‐scale studies due to its limitation to local measurements. As a result, in this study, we leverage spaceborne GNSS Reflectometry (GNSS‐R) developed as an innovative remote sensing technique. It exploits existing signals from numerous GNSS satellites after reflecting from the Earth's surface. The GNSS‐R satellites require only low‐cost, low‐mass, and low‐power receiver components, making them cost‐efficient and enabling the development of multi‐satellite constellations. GNSS‐R applications include monitoring of various environmental and atmospheric parameters such as soil Geophysical Research Letters 10.1029/2024GL111462 ASGARIMEHR ET AL. 2 of 11 19448007, 2024, 19, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024GL111462 by Csic Organización Central Om (Oficialia Mayor) (Urici), Wiley Online Library on [12/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License moisture (Camps, Park, et al., 2020; Rahmani et al., 2022), ocean surface, wind, and atmosphere conditions, for example, (Asgarimehr, Wickert, & Reich, 2018; Asgarimehr, Zavorotny, et al., 2018; Asgarimehr et al., 2021), ice and snow properties, for example, (Ghiasi et al., 2023; Hu et al., 2022; Song et al., 2022). Previous studies show the potential of GNSS‐R in monitoring of vegetation and biomass properties (Camps et al., 2016; Eroglu et al., 2019; Santi et al., 2021; X. Wu et al., 2021; Q. Yuan, Li, et al., 2019; Yueh et al., 2020) and here we will focus on retrieving VWC and studying plants' moisture dynamics in the Amazon. NASA's Cyclone GNSS (CYGNSS) constellation comprises eight microsatellites that can simultaneously track up to four reflected GNSS signals. As a result, CYGNSS achieves an unparalleled sampling rate of three to 7 hr (Ruf et al., 2018). Operating with 35°‐inclined orbits, CYGNSS provides frequent measurements over tropical latitudes at different times of the day. This advantage sets it apart from conventional radars, usually in sun‐ synchronous orbits, which offer observations with a much longer revisit time, always at the same time of day. These unique attributes of CYGNSS data make it possible to reconstruct and study the diurnal VWC cycles in the Amazon. Signals scattered from vegetated areas are sensitive to the dielectric properties of the soil, governed by soil moisture, as well as plant water content, influenced by leaf area and vegetation moisture (Ulaby et al., 1982). The VWC parameter represents the combined effect of the vegetation parameters. Our simulation study shown in Figure 1, confirms that L‐band reflectivity extracted from GNSS‐R measurements over forests primarily depends on VWC rather than soil moisture. While soil moisture affects VWC, its direct impact on GNSS‐R reflectivity is more pronounced in open areas or regions with sparse vegetation cover. As the incidence angle increases, GNSS signals traverse longer paths within vegetation layers, interacting with more vegetation volume, amplifying sensitivity to VWC. Due to the lower sensitivity to soil moisture, integrating soil moisture data from Soil Moisture Active Passive (SMAP) as ancillary data enables VWC retrieval from CYGNSS measurements, despite lower SMAP sampling frequency. While GNSS‐R offers high‐frequency sampling, its accuracy can be affected by variations in signal noise and the transmitted signal power from GNSS. The retrieval process depends on the direct signal power, which can vary between GPS blocks and over time. Understanding and accounting for these Figure 1. Simulated GNSS‐R effective reflectivity over Evergreen broadleaf forests at incidence angles θof 40°(a) and 70° (b). The white shades around the lines show the soil moisture impact as the standard deviation of reflectivity found by Monte Carlo simulation with 6% soil moisture error. See Supporting Information S1 for details. Geophysical Research Letters 10.1029/2024GL111462 ASGARIMEHR ET AL. 3 of 11 19448007, 2024, 19, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024GL111462 by Csic Organización Central Om (Oficialia Mayor) (Urici), Wiley Online Library on [12/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License variations is important for accurately interpreting GNSS‐R retrievals. The information on the power of transmitted signal is currently available from the CYGNSS data set and noise level can be extracted from measurements. For retrieval details and data sets, refer to the Supplement. 3. Results 3.1. VWC Dynamics in Amazon A VWC data set was compiled from CYGNSS measurements spanning 2019–2021. Figure 2shows the VWC across the study area, normalized by dividing it by the leaf area index (LAI), obtained from ERA5‐land reanalysis estimates, to account for seasonality and changes in biomass. This normalization removes the influence of biomass variations, focusing instead on the water content per unit leaf area, providing a more accurate representation of the plant's water status. Normalization using LAI has been previously proposed at the leaf level to assess how water content changes per unit leaf area across various species (F. Zhang & Zhou, 2019). Higher Normalized VWC (NVWC) values are observed during wet seasons, with a noticeable reduction during dry seasons. This pattern is consistent with previous studies reporting similar trends in vegetation moisture during dry periods due to reduced precipitation and increased atmospheric demand for water, for example, (van Emmerik et al., 2017; H. Wang et al., 2023). The seasonal cycles of NVWC are effectively captured by the GNSS‐R measurements, as illustrated in Figure 2g. These measurements demonstrate consistency with the VWC data from SMAP, as well as with the monthly average precipitation estimates from the ERA5 reanalysis. GNSS‐R reveals short‐term scale cycles with greater clarity, thanks to its higher temporal resolution. It also captures diurnal cycles, which are not detected by SMAP, which will be discussed in Figure 3. The spatial patterns of VWC in Figure 2demonstrate variability across the Amazon, influenced by local environmental microclimate conditions such as precipitation, soil type and moisture, temperature and vegetation cover. The derived NVWC also reveals year‐to‐year changes. The NVWC in the 2020 wet season is slightly higher compared to other years, consistent with the increased precipitation in this season. Weather station data also report higher precipitation during this period. The annual total rainfall divided by wet days in 2020 was 16.2 mm/day, compared to 15.4 mm/ day in both 2019 and 2021 (de Bodas Terassi et al., 2024). Figure 3presents the diurnal vegetation moisture cycles derived from GNSS‐R reflectivity observations over Amazon evergreen forests within the area of interest. This figure indicates that VWC exhibits significant diurnal fluctuations due to the processes of water uptake and transpiration. During the day, transpiration causes a decrease in VWC, which is replenished at night (Nelson et al., 2018). The higher VWC level compared to the evening value in the morning is evident in wet seasons 2019, 2020, by ≈1 and 0.6 kg/m2respectively. Dry seasons 2019 and 2021 show a similar pattern with discrepancies of ≈1.1 kg/m2. The observed cycles during dry periods show a more pronounced decrease in the afternoon VWC compared to the wet season. This is primarily due to the significantly higher VPD, which lead to increased transpiration rates. However, in water stress cases where water availability is limited, the transpiration rate drops, as discussed in the following section. In the dry seasons of 2020 and 2021, no clear pattern emerges, prompting the need to elucidate how atmospheric drivers govern these patterns. It is worth noting that the unclear patterns may stem from the heterogeneity of plants and their varied diurnal behaviors, potentially resulting in obscured cycles upon averaging. As a solution, the subsequent section provides a more targeted investigation at a smaller scale to study how diurnal VWC dynamics respond to water stress. 3.2. Response to Water Stress—Amazonian Peatland We investigate the GNSS‐R VWC observations at the AmeriFlux site PE‐QFR (3°50′03.9″S; 73°19′08.1″W) located in the Peruvian Amazon. This allows us to explore the diurnal behavior of VWC and its response to the parameters recorded by this Flux station, situated within a protected natural palm swamp peatland forest. The dominant tree species in this area are Mauritia flexuosa palms (61%) and Tabebuia insignis (15%) (F. Yuan et al., 2023). Figure 4shows the time‐series of observed ΔVWC =VWCpm −VWCam at a 0.25°proximity to the flux station, along with those of VPD, precipitation, and soil temperature, measured by the flux station. VWCpm −VWCam are obtained as 2‐hr average values centered at 5 p.m. and 5 a.m. Figures are smoothed using the method given by Eilers (2003). Thanks to the CYGNSS high sampling rate, the ΔVWC is reconstructed using a 7‐day moving time Geophysical Research Letters 10.1029/2024GL111462 ASGARIMEHR ET AL. 4 of 11 19448007, 2024, 19, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024GL111462 by Csic Organización Central Om (Oficialia Mayor) (Urici), Wiley Online Library on [12/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Figure 2. Average Normalized Vegetation Water Content (VWC) during wet (December–June) and dry (July–November) seasons for the years 2019 (a, b), 2020 (c, d), and 2021 (e, f). The maps are created in a 36 km grid. Panel (g) displays the time series of average Normalized VWC from GNSS‐R and SMAP, along with the monthly precipitation data from ERA5 reanalysis. The background colors indicate the dry and wet seasons, with dry periods shown in yellow and wet periods in blue. Geophysical Research Letters 10.1029/2024GL111462 ASGARIMEHR ET AL. 5 of 11 19448007, 2024, 19, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024GL111462 by Csic Organización Central Om (Oficialia Mayor) (Urici), Wiley Online Library on [12/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License window. A predominantly negative value of ΔVWC is observed throughout most of the time, except during specific periods highlighted with a red background in the figure. These highlighted periods correspond to arid conditions characterized by elevated VPD. The peak of ΔVWC occurs in July, coinciding with the period of maximum VPD and minimum precipitation, suggesting intensified water stress during this time. Figure 3. The Vegetation Water Content (VWC) diurnal cycles in wet (December–June) and dry (July–November) seasons for the years 2019 (a, b), 2020 (c, d), and 2021 (e, f). Geophysical Research Letters 10.1029/2024GL111462 ASGARIMEHR ET AL. 6 of 11 19448007, 2024, 19, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024GL111462 by Csic Organización Central Om (Oficialia Mayor) (Urici), Wiley Online Library on [12/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 3.3. VPD Is a Dominant Driver of Diurnal VWC Cycle In Figure 4, a notable observation is the strong similarity between ΔVWC and VPD, with a correlation coefficient of R=0.8. This association between VPD and VWC cycles is further highlighted in Figure 5, where a disrupted VWC cycle is consistently seen in the third quantile of VPD. Conversely, in the first and third quantiles of VPD, we observe the typical pattern of higher VWC levels throughout the year. Figure 5also demonstrates that abundant precipitation does not necessarily guarantee a healthy VWC cycle, neither in the dry nor wet season. Similarly, low soil temperatures may cause higher morning VWC levels in the wet season, but not in the dry season. Thus, our analysis suggests that ΔVWC is strongly influenced by VPD, particularly in the Amazonian peatland. VPD significantly influences dryness, vegetation water stress, and widespread tree mortality (Allen et al., 2010). By affecting the rate of transpiration, VPD directly impacts vegetation moisture. Elevated VPD increases evaporative demand due to the substantial difference between leaf water vapor and the surrounding air. In water deficit conditions, high VPD disrupts the plant moisture diurnal cycle, as plants adapt by closing their stomata during periods of high VPD to mitigate water loss (Oren et al., 1999). Stomatal closure reduces transpiration rates and plants continue water uptake to replenish themselves and compensate for the remaining low rates of water loss. This can in extreme cases appear as higher conserved moisture during the day compared to night. Midday stomatal closure is common and an observed in different forests (T. Brodribb & Holbrook, 2004; Brum et al., 2023; Konings et al., 2017; Pons & Welschen, 2003) as well as Eastern Amazon (Fisher et al., 2006). Research on tropical cloud forest trees and Platycladus orientalis has shown that high VPD is also associated with significant reductions in sap flow rates and greater maximum daily stem shrinkage (Brum et al., 2023). These regulatory responses imbalance the day water loss and night uptake rates. Finally, our study shows neither precipitation nor soil temperature emerges as the sole driver of water stress. Instead, it is the combined effect of these factors, as represented by VPD, that plays a crucial role in inducing water stress. This finding is consistent with recent studies emphasizing atmospheric water demand as the primary Figure 4. Time series of ΔVWC =VWC pm ‐ VWCam (a) obtained from GNSS‐R as well as VPD (b), precipitation (c), soil temperature (d), measured by the Flux station. The periods with ΔVWC >0 are highlighted with red background. Geophysical Research Letters 10.1029/2024GL111462 ASGARIMEHR ET AL. 7 of 11 19448007, 2024, 19, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024GL111462 by Csic Organización Central Om (Oficialia Mayor) (Urici), Wiley Online Library on [12/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License driver of water stress (Novick et al., 2016; Sulman et al., 2016). Although soil moisture is also recognized as a crucial factor inducing stress in semi‐arid ecosystems (L. Liu, Gudmundsson, et al., 2020), our study lacks access to in situ soil moisture data in the region for an evaluation in the Amazon, thus leaving this aspect for future investigation. 4. Conclusion—A New Avenue Understanding how tropical vegetation responds to water stress is crucial to assess the role of forested regions in the carbon budget within a changing climate. To achieve this, a remote sensing information source has been sought to fill in the data gaps left by traditional measurements of soil and plant water potential. Progress has been limited due to the inadequate sampling rate of conventional satellite sensors, making it challenging to effectively capture the diurnal variation in VWC, which, as found by this study, serves as a reliable indicator of vegetation moisture conditions. The introduction of GNSS‐R measurements can revolutionize the field by offering an unprecedented sampling rate that captures and maps the diurnal VWC cycles for the first time. The use of CYGNSS GNSS‐R observations has provided clear VWC mappings over the Amazon, allowing us to observe the diurnal dynamics of vegetation Figure 5. Vegetation Water Content (VWC) in the morning (green) and evening (blue), along with the difference ΔVWC =VWC pm −VWCam (red), shown at three quantiles of Vapor Pressure Deficit (VPD), precipitation, and soil temperature. The quantiles divide the data sets into three evenly distributed segments. The analysis is carried our in segregating cases January–March, April–June, July–September, and October–December due to the seasonal changes. Geophysical Research Letters 10.1029/2024GL111462 ASGARIMEHR ET AL. 8 of 11 19448007, 2024, 19, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024GL111462 by Csic Organización Central Om (Oficialia Mayor) (Urici), Wiley Online Library on [12/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License responses to water stress. In the Amozonian peatland, these cycles are predominantly influenced by VPD, indicating the significant role of this parameter as a water stress driver. The VWCs derived from GNSS‐R measurements exhibit consistency with those obtained from the SMAP data set and also show a geophysically plausible response to site‐measured VPD and water stress. To enhance accuracy and calibration of VWC GNSS‐R estimations, future studies should consider establishing ground‐based stations for measuring vegetation and soil moisture as well as varied atmospheric parameters. The current availability of ground data is highly limited in the Amazon, underscoring the need for new stations and comprehensive field campaigns to gather extensive moisture measurements. It is essential to acknowledge that the response of vegetation to water stress varies depending on factors such as vegetation types, biogeographical parameters, and broader environmental gradients. Therefore, the analysis of the Amazonian peatland cannot be directly generalized to other regions with different climates and vegetation types. Future studies are required to model water potential (ψ)as a function of VWC dynamics so that the derived information can be incorporated into existing hydraulic models. The availability of large VWC data sets from such novel sensors also paves the way for leveraging deep learning to model and study forests' responses to climate change and intensified droughts on a global scale. The provision of more extensive and frequent data from commercial GNSS‐R constellations will augment the potential. Looking ahead, the forthcoming ESA HydroGNSS mission promises to enhance the GNSS‐R data set with unique measurement characteristics such as in different polarizations and using the signals from other GNSS constellations. Data Availability Statement All data used in this study are publicly accessible; The VWC retrieval in this study uses CYGNSS Level 1 version 3.0 data, accessible via CYGNSS (2020). As ancillary data to compute the Fresnel reflection coefficient, we used SMAP L2 Radiometer 36 km EASE‐Grid Soil Moisture (SPL2SMP), Version 6 data, retrievable from O’Neill et al. (2019), and the SMAP L3 Radar/Radiometer Global Daily 9 km EASE‐Grid Soil Moisture (SPL3SMAP), Version 3, accessible at Entekhabi et al. (2016). Additionally, we acquired the clay fraction from the Global Land Data Assimilation System (GLDAS) map, provided by Rodell et al. (2004). Rain‐contaminated measurements are identified and excluded from the analysis using the integrated multi‐satellite retrievals of Global Precipitation Measurement (IMERG) final precipitation L3 product, available at Huffman et al. (2019). Furthermore, land cover classification gridded maps are obtained from satellite observations and accessible at Copernicus Climate Change Service (2019), to distinguish forests from other land covers in our analysis. The data from the AmeriFlux site PE‐QFR, including VPD, soil temperature, and precipitation, are available at Roman et al. (2021). References Allen, C. D., Macalady, A. K., Chenchouni, H., Bachelet, D., McDowell, N., Vennetier, M., et al. (2010). A global overview of drought and heat‐ induced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management,259(4), 660–684. https://doi.org/10. 1016/j.foreco.2009.09.001 Asgarimehr, M., Hoseini, M., Semmling, M., Ramatschi, M., Camps, A., Nahavandchi, H., et al. (2021). Remote sensing of precipitation using reflected GNSS signals: Response analysis of polarimetric observations. IEEE Transactions on Geoscience and Remote Sensing,60, 1–12. https://doi.org/10.1109/tgrs.2021.3062492 Asgarimehr, M., Wickert, J., & Reich, S. (2018). TDS‐1 GNSS reflectometry: Development and validation of forward scattering winds. Ieee Journal of Selected Topics in Applied Earth Observations and Remote Sensing,11(11), 4534–4541. https://doi.org/10.1109/jstars.2018. 2873241 Asgarimehr, M., Zavorotny, V., Wickert, J., & Reich, S. (2018). Can GNSS reflectometry detect precipitation over oceans? Geophysical Research Letters,45(22), 12–585. https://doi.org/10.1029/2018gl079708 Bernardino, P. N., Oliveira, R. S., Van Meerbeek, K., Hirota, M., Furtado, M. N., Sanches, I. A., & Somers, B. (2024). Estimating vegetation water content from Sentinel‐1 C‐band SAR data over savanna and grassland ecosystems. Environmental Research Letters,19(3), 034019. https://doi. org/10.1088/1748‐9326/ad288f Brodribb, T., & Holbrook, N. (2004). Diurnal depression of leaf hydraulic conductance in a tropical tree species. Plant, Cell and Environment, 27(7), 820–827. https://doi.org/10.1111/j.1365‐3040.2004.01188.x Brodribb, T. J., Powers, J., Cochard, H., & Choat, B. (2020). Hanging by a thread? Forests and drought. Science,368(6488), 261–266. https://doi. org/10.1126/science.aat7631 Brum, M., Vadeboncoeur, M., Asbjornsen, H., Puma Vilca, B. L., Galiano, D., Horwath, A. B., & Metcalfe, D. B. (2023). Ecophysiological controls on water use of tropical cloud forest trees in response to experimental drought. Tree Physiology,43(9), 1514–1532. https://doi.org/10. 1093/treephys/tpad070 Camps, A., Alonso‐Arroyo, A., Park, H., Onrubia, R., Pascual, D., & Querol, J. (2020). L‐band vegetation optical depth estimation using transmitted GNSS signals: Application to GNSS‐reflectometry and positioning. Remote Sensing,12(15), 2352. https://doi.org/10.3390/ rs12152352 Acknowledgments The work leading to this publication was supported by the Postdoctoral Researchers International Mobility Experience (PRIME) programme of the German Academic Exchange Service (DAAD) with funds from the German Federal Ministry of Education and Research (BMBF). The authors express gratitude to Brid Schenkl and the PRIME team at DAAD for their assistance in project administration, ensuring its success. Open Access funding enabled and organized by Projekt DEAL. Geophysical Research Letters 10.1029/2024GL111462 ASGARIMEHR ET AL. 9 of 11 19448007, 2024, 19, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024GL111462 by Csic Organización Central Om (Oficialia Mayor) (Urici), Wiley Online Library on [12/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License